Incorporating Topographic Stress Perturbations and Geologic Structures into Roof Stability Forecasting in Underground Mines (PI: W. Ashley Griffith, University of Akron) Project Summary Some studies indicate that within the northern Appalachian Plateau physiographic province (roughly speaking, eastern Ohio, western Pennsylvania, and northern West Virginia), where underground coal mining is extremely prevalent, approximately 90% of roof falls occur in mines beneath stream valleys (Moebs and Stateham, 1984; 1985). Roof collapse has been studied for over a century, with resulting standard practices for dealing with roof instabilities based on empirical or back-of the envelope calculations (e.g., Mucho and Mark, 1994). These practices typically either ignore the effects of stress perturbations due to topographic relief at the earth's surface. Pre-existing geologic structures such as faults, joints, and buried stream channels may also cause stress perturbations and/or form surfaces of weakness. Such geologic structures are frequently accounted for, although often only qualitatively. However, recent advances in Light Detection and Ranging (LiDAR) scanning to reconstruct large scale three dimensional surfaces as well as easy-to-use, fast mechanical computer models for calculating stress-states due to three-dimensional structures should make it possible to take these complexities into account in an manner sufficiently efficient to influence mine construction decisions as the mine is being excavated. Here we propose to test this hypothesis on an active shallow (200-500 ft deep) coal mine in eastern Ohio, where rock falls have been un-predictable by standard practice measures, and relief in surface topography is of the same scale as the depth of the mine. The study site, the Carroll Hollow Mine, is operated by the Sterling Mining Corporation, the underground mining subsidiary of the East Fairfield Coal Company, and work will be conducted in conjunction with Tim Miller, the mine geologist. A 3D model surface, consisting of a grid of triangular elements, will be created from a freely available LiDAR point cloud of the topography in the vicinity of the mine, augmented by differential geographic positioning system (D-GPS) surveys by the PI and graduate student in areas where the surface topography has been altered due to surface mining. The three dimensional stress state due to gravitational perturbations around this topographic surface as well as far-field tectonic stresses constrained by direct measurements of stress in boreholes will be calculated throughout the existing mine and planned excavations using the numerical boundary element method code Poly3D. Geologic structures and roof falls will be mapped throughout the mine in an AutoCAD database, and compared to the stress field calculated numerically. In subsequent model iterations, increasing complexity (faults, excavation geometry) will be incorporated into the numerical model until the theoretical 3D stress state best matches the observed rock fall distribution. In the process, an algorithm will be created to help future mine operations incorporate topography into roof stability forecasting during mine planning. Results will be presented at national scientific meetings, and published in two peer-reviewed publications, one focusing on the site-specific study, and the other focusing on how our work flow can be adapted to mine planning strategies.